In Harvard’s Admissions Decisions, Signs of Progress but Some Data Missing

While many were upset when the U.S. Supreme Court curtailed Harvard’s use of racial preferences, I suggested that the decision might eventually serve working-class students of all races.

For decades, Harvard has brought together wealthy students of all colors, but limiting the use of race could prompt the University to reach out to more low-income and working-class students, indirectly creating racial diversity without violating the law.

When Harvard announced its admissions decisions for the Class of 2028 last Thursday, it provided some evidence that it is headed in the right direction. Although it disclosed no racial data and only scant information about socioeconomic status, one number suggests continual progress in an important area: the percentage of first-generation students.

According to data released Thursday, the percentage of first-generation college students for Harvard’s Class of 2028 has risen to 20.5 percent, nearly tripling the seven percent representation of first-generation students Harvard admitted in the Class of 2019. Last year, Harvard reported this number being nearly 20 percent.

Despite impressive long-term progress, Harvard will need to push even further if it wants to draw upon talents from students of all socioeconomic backgrounds, preserving racial diversity. Harvard should continue to diversify by metrics such as the education level of parents, income level, and other indicators of socioeconomic status.

As an expert witness in the affirmative action litigation, I worked with Duke economist Peter S. Arcidiacono to model what Harvard could do to ensure high levels of racial and economic diversity and maintain superb academic standards.

In the simulations of workable plans to achieve racial diversity without racial preferences, we demonstrated that Harvard’s first-generation student numbers could increase to 25 percent — substantially higher than Harvard’s numbers for the Class of 2028. (The simulations also assumed that Harvard would end preferences for legacies and children of faculty, something the University has not announced.)

The other socioeconomic data Harvard released on Thursday is of limited utility.

Harvard said that 20.7 percent of the admitted Class of 2028 is eligible for federal Pell grants, for which families in roughly the bottom socioeconomic half qualify. While significant, scholars have shown that Pell data can be a misleading proxy, since it includes many students that are squarely middle-class, just within the range of eligible family incomes.

As a result, Pell rates may not track genuine representation of low-income students. For example, Paul Tough concluded in his book “The Years That Matter Most: How College Makes or Breaks Us,” that Princeton University doubled its Pell representation from 2004-2013 “while only barely increasing the number of actual low-income students on its campus.”

Moreover, Pell data also tells us nothing about the socioeconomic distribution of the more than three-quarters of the admitted class of 2028 who are above the Pell cutoff.

To fill this gap, David J. Deming, a Harvard professor, has argued that colleges should report more detailed income data, including the share of students coming from the top one percent. Raj Chetty ’00, another Harvard professor, has broken down Harvard’s income data by quintile for the Class of 2013, but I believe this information should be reported annually.

I would even go further, requiring colleges to provide a breakdown of student body populations by wealth — that is, net worth — and by neighborhood poverty levels.

Wealth and neighborhood poverty are significant predictors of opportunity in America, and Black and Hispanic students are much more likely to be disadvantaged than white students from the same income level. Harvard should be rewarding talented students who overcome these obstacles, and without the data, we cannot know whether Harvard is doing so.

Harvard should also release the number of students who are legacies or the children of faculty and staff.

Harvard’s expert witness, professor David E. Card, showed that under the simulations of economic affirmative action in which preferences for legacies and faculty children were removed, the percentage of legacy students in the class of 2019 would have dropped by nearly 70 percent. The share of faculty and staff children would have dropped by a slightly lower proportion. Release of data would show the extent to which preferences remain.

Despite its shortcomings, the data Harvard did report are encouraging. The College deserves credit for significantly increasing the share of first-generation college students during the legal fight over the use of race in admissions.

Still, as an institution whose motto is Veritas, Harvard should be willing to lift the veil even further.

Richard D. Kahlenberg is the director of the American Identity Project at the Progressive Policy Institute. He was an expert witness for Students for Fair Admissions in its lawsuit against Harvard.

This piece is the second installment in a series that will provide analysis and commentary on the culmination of Harvard College’s first admissions cycle following the United States Supreme Court’s curtailment of race-conscious admissions. Read Kahlenberg’s first installment here.